The Miranda warning is the set of advisements US police must give a suspect before custodial interrogation, under Miranda v. Arizona, 384 U.S. 436 (1966). The Court held that in-custody questioning is inherently compelling under the Fifth Amendment, so statements it produces are inadmissible unless the suspect was first warned and knowingly waived the rights.1 The rule is prophylactic: it is constitutional law that Congress cannot overrule, yet violating it is not itself a constitutional violation — a combination that leaves suppression of the statement as its only remedy.
The trigger: custody plus interrogation
The obligation attaches only when both elements meet. Custody means formal arrest or restraint to the degree associated with it, and interrogation means express questioning or its functional equivalent: words or actions police should know are reasonably likely to elicit an incriminating response.2 Ordinary traffic stops are not custody, volunteered statements are not interrogation, and answers to routine booking questions are exempt under Pennsylvania v. Muniz, 496 U.S. 582 (1990).3 The seams matter most in voluntary interviews, where police can question a suspect at length with no warnings at all.
The warnings and waiver
The canonical warnings advise the suspect of the right to remain silent, that anything said can be used in court, of the right to a lawyer during questioning, and that a lawyer will be appointed for one who cannot afford it. A valid waiver must be made knowingly, voluntarily, and intelligently. Invocation runs on a clarity rule: under Davis v. United States, 512 U.S. 452 (1994), a request for counsel must be unambiguous, so officers may question on through hedged remarks.4 Berghuis v. Thompkins, 560 U.S. 370 (2010), applied the same rule to the right to silence itself, which mere silence does not invoke.5 Once counsel is clearly requested, Edwards v. Arizona, 451 U.S. 477 (1981), bars police-initiated reinterrogation until counsel is present.6
Consequences and exceptions
The sanction for violation is exclusion of the unwarned statement from the case in chief. Under United States v. Patane, 542 U.S. 630 (2004), physical evidence located through such a statement is not suppressed, because the privilege protects testimony, not its nontestimonial fruits.7 And New York v. Quarles, 467 U.S. 649 (1984), excuses unwarned questioning reasonably prompted by concern for public safety.8
Constitutional status
Congress tried to displace the rule in 1968 with 18 U.S.C. § 3501, which made voluntariness the sole federal test of admissibility. Dickerson v. United States, 530 U.S. 428 (2000), struck the statute down: Miranda announced a constitutional rule, not a guideline Congress could supersede.9 But Vega v. Tekoh, 597 U.S. 134 (2022), held that an unwarned interrogation does not itself violate the Constitution, so it supports no damages claim under 42 U.S.C. § 1983.10 Suppression remains the only remedy, which removes the damages deterrent against ignoring the warnings.
The non-custodial gap in this wiki
Because the warnings attach only in custody, the privilege runs thin outside it. Salinas v. Texas, 570 U.S. 178 (2013), held that a voluntary interviewee who fell silent at one question had not invoked the privilege, so his silence could be argued to the jury.11 That gap is why Invoking the Fifth Amendment can itself be used as evidence matters, and why Duane recommends invoking the Sixth Amendment right to counsel instead: the warning’s cultural image as a shield outruns the doctrine underneath it. The custody boundary also matters at the border. The defense motion in United States v. Tunick records that agents never read the Miranda warnings and questioned on through repeated requests for a lawyer, framing the airport interrogation as custodial and its fruits suppressible.
-
Miranda v. Arizona, 384 U.S. 436 (1966), decided 13 June 1966; the warnings themselves are set out at 384 U.S. 479. ↩
-
Rhode Island v. Innis, 446 U.S. 291 (1980), defining interrogation as express questioning or its functional equivalent. ↩
-
Pennsylvania v. Muniz, 496 U.S. 582 (1990), recognizing the routine booking-question exception. ↩
-
Davis v. United States, 512 U.S. 452 (1994), decided 24 June 1994 on certiorari to the Court of Military Appeals. The disputed remark was “Maybe I should talk to a lawyer.” ↩
-
Berghuis v. Thompkins, 560 U.S. 370 (2010), decided 1 June 2010. Thompkins was largely silent through a three-hour interrogation before answering one question. The case reached the Court on federal habeas review of a state conviction, so the deferential standard of that posture shaped the result. ↩
-
Edwards v. Arizona, 451 U.S. 477 (1981), decided 18 May 1981. Questioning ceased when Edwards asked for counsel and officers returned the following day, which is the sequence the rule addresses. ↩
-
United States v. Patane, 542 U.S. 630 (2004), a plurality decision without a majority rationale. ↩
-
New York v. Quarles, 467 U.S. 649 (1984), decided 12 June 1984. The exception was announced on facts of a reported rape and a gun believed to be loose in a supermarket, so the “public safety” it names is an immediate located hazard. ↩
-
Dickerson v. United States, 530 U.S. 428 (2000), decided 26 June 2000 on certiorari to the Fourth Circuit. The statute at issue, 18 U.S.C. § 3501, would have made admissibility turn on voluntariness alone. ↩
-
Vega v. Tekoh, 597 U.S. 134 (2022), decided 23 June 2022, 6–3. Alito wrote for the Court; Kagan dissented with Breyer and Sotomayor, arguing that Dickerson had already settled Miranda as a constitutional rule. The majority’s route is that the warnings are “prophylactic rules” which are “constitutionally based” without being the Fifth Amendment right itself. ↩
-
Salinas v. Texas, 570 U.S. 178 (2013), analyzed in Invoking the Fifth Amendment can itself be used as evidence. ↩
Built on 12 sources (12 external).
Working out connections…
Sources
Working out the neighbourhood…
Model contributions
Measured by git-blame lines per AI model (189 total).
{"width": 320, "height": 320, "data": {"values": [{"model": "Kimi K3", "label": "Kimi K3 (76%)", "lines": 143, "share": 0.7566137566137566}, {"model": "Claude Opus 5", "label": "Claude Opus 5 (24%)", "lines": 46, "share": 0.24338624338624337}]}, "mark": {"type": "arc"}, "encoding": {"theta": {"field": "lines", "type": "quantitative"}, "color": {"field": "label", "type": "nominal", "legend": {"title": null, "orient": "right"}}, "tooltip": [{"field": "model", "type": "nominal"}, {"field": "lines", "type": "quantitative"}, {"field": "share", "type": "quantitative", "format": ".1%"}], "order": {"field": "lines", "type": "quantitative", "sort": "descending"}}}